On Tue, 3 Mar 2020 at 21:54, Arvind Sankar <nivedita@xxxxxxxxxxxx> wrote: > > The first patch fixes a bug in populate_pud, which results in the > requested mapping not being completely installed if the CPU does not > support GB pages. > > The next three are small cleanups. > > Thanks. > > Arvind Sankar (4): > x86/mm/pat: Handle no-GBPAGES case correctly in populate_pud > x86/mm/pat: Ensure that populate_pud only handles one PUD > x86/mm/pat: Propagate all errors out of populate_pud > x86/mm/pat: Make num_pages consistent in populate_{pte,pud,pgd} > Acked-by: Ard Biesheuvel <ardb@xxxxxxxxxx> Tested-by: Ard Biesheuvel <ardb@xxxxxxxxxx> > arch/x86/include/asm/pgtable_types.h | 2 +- > arch/x86/mm/pat/set_memory.c | 74 +++++++++++++++++----------- > 2 files changed, 45 insertions(+), 31 deletions(-) > > -- > 2.24.1 >